This book presents a concise introduction to Bartlett and Bartlett-type corrections of statistical tests and bias correction of point estimators. The underlying idea behind both groups of corrections is to obtain higher accuracy in small samples. While the main focus is on corrections that can be analytically derived, the authors also present alternative strategies for improving estimators and tests based on bootstrap, a data resampling technique and discuss concrete applications to several important statistical models.

Provides a unified overview of Bartlett corrections and bias reduction Discusses bootstrap-based inference Includes applications to important statistical models Includes supplementary material: sn.pub/extras

Autorentext

Gauss M. Cordeiro is a Professor of Statistics at Universidade Federal de Pernambuco in Brazil. He is a former chief editor of the Brazilian Journal of Probability and Statistics and a former president of the Brazilian Statistical Association.

Francisco Cribari-Neto is a Professor of Statistics at Universidade Federal de Pernambuco in Brazil. He is a former applications editor of the Brazilian Journal of Probability and Statistics and a former president of the Brazilian Econometric Society.

Inhalt
Preface.- Likelihood-Based Inference and Finite-Sample Corrections: A Brief Overview.- Bartlett Corrections and Bootstrap Testing Inference.- Bartlett-Type Corrections.- Analytical and Bootstrap Bias Corrections.- Supplementary Material.- Glossary.